ANIMAL NATURE OF DIATOME/E. 417 



power do not exceed GOO diameters, and by causing the 

 image. from the camera lucida to fall upon it, I have its 

 measurement immediately taken. I have constructed 

 one of these scales for every combination of my micro- 

 scope, and thus by a simple application of the camera 

 lucida, I can measure every object without at all changing 

 the conditions of the observation. I dwell upon the 

 method I pursue in my microscopical researches, to prove 

 that I devote to it scrupulous accuracy. The screw 

 micrometer also gives millimillimetres, and by the addi- 

 tion of a nonius even decimihimillimetres ; but besides 

 that 'it is inconvenient to keep it applied to the plate of 

 the stage, and that there is great loss of time and inter- 

 I'uption of the observation in applying it when its use 

 is required, it has always the great iuconvenience of waste 

 of trouble. In ten observations with the screw micro- 

 meter, we scarcely find two that perfectly agree. With 

 the camera lucicla one only is sufficient. I think it 

 unnecessary to bring arguments against the method 

 of applying a glass micrometer over every object we 

 wish to measure. Still the plan of a micrometer fixed 

 in the eye-glass, and previously corrected by examining 

 another micrometer as the object of observation, pos- 

 sesses great convenience, though not perhaps scrupu- 

 lous accuracy. But we have not merely to measure the 

 objects we are about to describe ; we must also define 

 that measurement. It would seem so simple a thing for 

 all to use the same standard, and there is so much con- 

 venience in the metrical measure and decimal notation, 

 that it excites real wonder to see how prevalent among men 

 of science is the habit of preferring the duodecimal mea- 

 surement peculiar to every country, and expressing that 

 in vulgar, not decimal fractions. The evil would be less 

 were any one (standard) adopted, constantly used, and 

 defined. But the matter is worse than this. Ehrenberg 

 speaks perpetually of a line, without stating the standard ; 

 at the same time he gives its equivalent in metrical 

 admeasurement, and from this it appears that his line is 



37 



